Project Summary The proposed project will significantly advance the understanding of personal (family and friendship) network influences on adult alcohol use by using social network analysis to investigate how personal networks are related to patterns of drinking over time. Peer influence plays a central role in theories of problem behavior in adolescents, and has been identified as a key driver of adolescent drinking across a number of studies, but to date there is no nationally representative study of adult social networks and substance use. People born in recent years tend to drink more than older generations, suggesting that as the population ages, these individuals will continue to drink more ? particularly important given that adults age 60+ will comprise a quarter of the total U.S. population by 2040 (up from one fifth today). The fields of alcohol research and prevention may see substantial benefit from the use of novel methodological techniques to develop models that may provide a clearer understanding of the ways in which social relationships influence (and are influenced by) adult drinking. Specifically, we will 1) compare the content and structure of the social networks of a nationally representative sample adult in two age groups: middle adulthood (ages 30-59) and older adulthood (age 60-80); 2) test for associations between the alcohol use behavior of adults and characteristics of their social networks, identifying distinct influence and selection effects in trajectories of alcohol use and AUD outcomes over time; 3) examine substance use disparities by sex, race/ethnicity and economic status over time and across stages of the lifespan. To do so, we will draw field social network and substance use surveys to a national probability sample of 1,700 individuals, conducting longitudinal surveys one year apart for four years. The study will provide new insights into the role of personal networks in adult alcohol use, and in disparities in health and practical information concerning ways to reduce them. As an additional benefit, data from the study will be publicly available and other researchers will also be able to continue to survey the panel, providing an ongoing resource to further scientific understanding of adult social networks. Thus, this study provides the initial framework for what could be one of the single most important longitudinal network analysis data sets available. This project is envisioned to form the foundation for a vast array of subsequent studies that seek to link longitudinal network data to individual behavior change among adults in numerous fields and disciplines